Ex Parte ShihadehDownload PDFBoard of Patent Appeals and InterferencesFeb 24, 201111098153 (B.P.A.I. Feb. 24, 2011) Copy Citation UNITED STATES PATENT AND TRADEMARK OFFICE ____________ BEFORE THE BOARD OF PATENT APPEALS AND INTERFERENCES ____________ Ex parte ELIAS SHIHADEH ____________ Appeal 2009-009800 Application 11/098,153 Technology Center 2100 ____________ Before JOHN A. JEFFERY, ST. JOHN COURTENAY III, and THU A. DANG, Administrative Patent Judges. COURTENAY, Administrative Patent Judge. DECISION ON APPEAL1 1 The two-month time period for filing an appeal or commencing a civil action, as recited in 37 C.F.R. § 1.304, or for filing a request for rehearing, as recited in 37 C.F.R. § 41.52, begins to run from the “MAIL DATE” (paper delivery mode) or the “NOTIFICATION DATE” (electronic delivery mode) shown on the PTOL-90A cover letter attached to this decision. Appeal 2009-009800 Application 11/098,153 2 STATEMENT OF THE CASE This is an appeal under 35 U.S.C. § 134(a) from the Examiner’s final rejection of claims 1-20, which are all the claims pending in the application. We have jurisdiction under 35 U.S.C. § 6(b). We Reverse. Representative Claim 1. A method comprising: quantifying a number of periodic pulses occurring during an execution of one or more instructions by a processor; and adjusting an available depth of branch predictions based on a comparison of the number of periodic pulses to a predefined value. Examiner’s Rejection Claims 1-20 stand rejected under 35 U.S.C. § 103(a) as being unpatentable over Buyuktosunoglu (US 2002/0053038 A1) and Jain (US 5,553,255). Claims 1-8 ISSUE Based upon our review of the administrative record, we have determined that the following issue is dispositive in this appeal regarding claims 1-8: Appeal 2009-009800 Application 11/098,153 3 Under § 103, did the Examiner err in determining that the combination of Buyuktosunoglu and Jain would have taught or suggested adjusting an available depth of branch predictions based on a comparison of the number of periodic pulses to a predefined value, within the meaning of independent claim 1? FINDINGS OF FACT 1. Jain teaches a special purpose register, accessible to the programmer, that generates three node bits. (Col. 9 ll. 34-37). 2. The Examiner states that “changing the size of any one of the dispatch buffer, the branch queue or the user- programmable prefetch mode bits, or changing several of them in combination, would change the “available depth of branch predictions.” (Ans. 35). 3. The Examiner further contends that the branch prediction table, dispatch buffer, and/or branch queue and the value of the prefetch mode bits, determine the number of branch predictions that can be stored/held. Therefore, the “available depth of branch predictions” is “adjusted.” (Id.). 4. The Examiner interprets the claimed “available depth of branch predictions” as the number of branch predictions that can be held at one time. (Id.). Appeal 2009-009800 Application 11/098,153 4 ANALYSIS The Examiner acknowledges that the primary Buyuktosunoglu reference does not teach the disputed limitation. (Ans. 8-9). The Examiner looks to the secondary Jain reference for teaching or suggesting adjusting an available depth of branch predictions based on a comparison of the number of periodic pulses to a predefined value, within the meaning of independent claim 1. (Ans. 9-10) See also FF 4. Appellant contends that Jain (and therefore the cited combination of references) does not teach that the maximum number of unresolved branch instructions is adjusted based on any criteria. Instead, Jain actually teaches that the maximum number of unresolved branch instructions is configured based on a user-programmable special purpose register. Thus, further prefetching is prevented once a predetermined/programmed maximum number of unresolved branch instructions has been reached. (App. Br. 8). Based upon our review of the evidence, we find Appellant’s arguments persuasive. As pointed out by the Appellant, we agree that according to Jain, the depth of the unresolved branch instructions is not adjusted. Instead, the special purpose register predefines or sets (i.e., is user programmable) the maximum number of unresolved branch instructions (FF 1-3), and is not adjusted based upon a comparison, or any other criteria. Moreover, we observe that the language of claim 1 expressly requires that the available depth is adjusted based on a comparison of the number of periodic pulses to a predefined value. (Claim 1). We agree with Appellant that the cited references do not teach nor fairly suggest this limitation. Appeal 2009-009800 Application 11/098,153 5 Based on this record, we find the Examiner erred in rejecting independent claim 1. Accordingly, we reverse the Examiner’s rejection of claim 1, as well as associated dependent claims 2-8. Claims 9-13 We note that independent claim 9 recites a limitation similar to the comparison of the number of periodic pulses to a predefined value, as discussed supra regarding claim 1. Specifically, claim 9 recites that the available speculative branch prediction depth is adjusted based on the instruction execution efficiency. For the reasons discussed above regarding claim 1 and also for the reasons argued by Appellant (App. Br. 12-13), we reverse the Examiner’s rejection of independent claim 9 and associated dependent claims 10-13. Claims 14-20 Based upon our review of the administrative record, we have determined that the following issue is dispositive in this appeal regarding claims 14-20: ISSUE Under § 103, did the Examiner err in determining that the combination of Buyuktosunoglu and Jain would have taught or suggested a “depth adjustment module [that] is further operable to adjust the first value between cycles of instructions based on a comparison of a first number of pulses per instruction received from the shifter for a first cycle of instructions to a second number of pulses per instruction received from the shifter for a Appeal 2009-009800 Application 11/098,153 6 second cycle of instructions subsequent to the first cycle of instructions,” within the meaning of independent claim 14? (emphasis added). ANALYSIS We observe that independent claim 14 recites a depth adjustment module that performs an adjustment based upon a comparison of pulses, similar to our discussion supra regarding claim 1. For the reasons discussed above regarding claim 1 and also for the reasons argued by Appellant (App. Br. 15-17), we reverse the Examiner’s rejection of independent claim 14 and associated dependent claims 15-20. DECISION We reverse the Examiner’s § 103 rejection of claims 1-20. ORDER REVERSED pgc LARSON NEWMAN & ABEL, LLP 5914 WEST COURTYARD DRIVE SUITE 200 AUSTIN, TX 78730 Copy with citationCopy as parenthetical citation